Effect of peripheral prisms on blind side gaze scanning at street crossings by pedestrians with homonymous hemianopia

Abstract

Purpose : People with homonymous hemianopia (HH), who have half of their visual field blind, could benefit from wearing field expansion devices. We investigated the effects of peripheral prisms on their gaze scanning towards the side of the missing visual field (blind side) at street crossings.

Methods : Gaze movements were recorded for 14 HH pedestrians as they crossed multiple streets in a busy urban area at baseline when they first received the prisms, and after 1.5 months of acclimation and training using prisms (follow-up). Baseline visit also included a session without prisms. Gaze scans > 20° at street crossings were counted. Changes in the blind side scanning rate with respect to the body midline (SR; scans per minute) and scan magnitude (SM; angles in degrees) with prisms were analyzed.

Results : Baseline data without and with prisms included 845 and 903 scans at 136 and 150 street crossing instances totaling 55 and 53 minutes duration, respectively. At follow-up, 725 scans were recorded at 161 street crossing instances totaling 55 minutes with prisms. At baseline, SR with prisms (mean[95% confidence interval]; 18.6[15.6-21.5]) was significantly higher than without prisms (14.5[11.6-17.4], p<.001), but it decreased to about the same level at follow-up (14.6[10.4-18.8], p<.001). At baseline, 10 out of 14 participants had higher SR with prisms, and it declined from baseline to follow-up in 11 participants. At baseline, SM was not significantly different between with and without prisms conditions. However, SM with prisms increased significantly from baseline (41[37-44]) to follow-up (46[42-50], p<.001), as 11 out of 14 subjects showed an increase in SM with prisms from baseline to follow-up.

Conclusions : Since the prisms shift the visual direction of objects from the blind side to the seeing side, they are expected to affect the bottom-up gaze scanning mechanism. Pre-acclimation, participants presumably noticed more shifted objects and scanned more often towards the blind side, but possibly faced some challenges in precisely shifting gaze to the objects in the blind side. After acclimation, they did not need to scan for every object detected through the prisms, but they were more likely to make larger scans towards the blind side.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.

 

Mean and 95% CI (error bars) for blind side SR (left) and SM (right) for HH pedestrians at street crossings.